Volume 55, Issue 2, Pages 289-300 (July 2007) Induction of ΔFosB in the Periaqueductal Gray by Stress Promotes Active Coping Responses  Olivier Berton, Herbert E. Covington, Karl Ebner, Nadia M. Tsankova, Tiffany L. Carle, Paula Ulery, Akshay Bhonsle, Michel Barrot, Vaishnav Krishnan, Georg M. Singewald, Nicolas Singewald, Shari Birnbaum, Rachael L. Neve, Eric J. Nestler  Neuron  Volume 55, Issue 2, Pages 289-300 (July 2007) DOI: 10.1016/j.neuron.2007.06.033 Copyright © 2007 Elsevier Inc. Terms and Conditions

Figure 1 Escape Deficit and ΔFosB Accumulation in the vlPAG after IS (A) IS-induced escape deficits in mice. (Left) Escape latency across 15 escape trials. (Right) Distribution of the average latency during the last five trials and median split (red line) used to discriminate between Escape deficit versus No escape deficit groups. No-IS group, n = 7, IS group, n = 28, ∗p < 0.05 versus IS-escape deficit, #p < 0.05 versus no-IS. (B) ΔFosB accumulation in vlPAG. Top: (Left) Immunolabeling for ΔFosB and (right) tryptophan hydroxylase (TPH). Bottom: (Left) Overlay and (right) coronal view corresponding to AP coordinate −4.36 mm relative to bregma. Aq, aqueduct; DR, dorsal raphe; mlf, medial longitudinal fasciculus. (C) (Left) ΔFosB accumulates most intensely in the area corresponding to the rostral portion of the DR in IS-no escape deficit group (ANOVA significant interaction between IS and rostrocaudal level, F12,104 = 21.33, p < 0.001, ∗∗∗p < 0.001, ∗∗p < 0.01 versus IS-no escape deficit) and correlates negatively with the severity of the behavioral deficit (r = −0.69, p < 0.01, n = 18). Data are expressed as mean ± SEM. Neuron 2007 55, 289-300DOI: (10.1016/j.neuron.2007.06.033) Copyright © 2007 Elsevier Inc. Terms and Conditions

Figure 2 ΔFosB Accumulates in Substance P-Positive Neurons after IS (A) Relative distribution of ΔFosB-positive cells (visualized with DAB) and PPT-A mRNA (insert, silver grains) in the vlPAG at 10× magnification. The black arrow indicates the position of the aqueduct on each view. (B) (Left) Autoradiographs illustrating the relative distribution of SERT mRNA, GAD-67 mRNA, and PPT-A mRNA at the same rostrocaudal level. The thin black arrows indicate the position of the aqueduct on each view. (Right) Examples of views (100×) used to quantify mRNA grain density over ΔFosB-positive cells (black circles), ΔFosB-negative cells (white circles), and background (red circle). Arrowheads indicate cells which are either positive (black) or negative (white) for a given mRNA signal. (C) Proportion of ΔFosB-positive cells colabeled with each type of probe in the vPAG; note the major colocalization of ΔFosB-positive cells with PPT-A mRNA. (D) Density of PPT-A silver grains is decreased in ΔFosB-positive cells compared with ΔFosB-negative cells; ∗p < 0.05. Data are expressed as mean ± SEM. Neuron 2007 55, 289-300DOI: (10.1016/j.neuron.2007.06.033) Copyright © 2007 Elsevier Inc. Terms and Conditions

Figure 3 ΔFosB Regulates PPT-A Gene Expression and Stress-Induced Substance P Release (A) Viral overexpression of ΔFosB dose-dependently inhibits PPT-A gene promoter activity in luciferase reporter assays; ∗p < 0.05. (B) ChIP from vlPAG demonstrates the direct binding of ΔFosB (but not full-length FosB) to the PPT-A gene promoter in vivo, 24 hr after IS administration; ∗p < 0.05. (C) In vivo microdialysis in freely moving rats. Exposure to IS or forced swimming stress increases substance P release in the NAc, an effect blocked by previous bilateral infusion of HSV-ΔFosB in the vlPAG. ANOVA effect of time (F11,165 = 6.74, p < 0.0001) and interaction between virus and time (F11,165 = 2.07, p = 0.02), ∗p < 0.05 and ∗∗p < 0.01 from baseline, #p < 0.05 and ##p < 0.01 from LacZ (n = 8–9 rats per group). Data are expressed as mean ± SEM. Neuron 2007 55, 289-300DOI: (10.1016/j.neuron.2007.06.033) Copyright © 2007 Elsevier Inc. Terms and Conditions

Figure 4 NK1 Receptor Blockade in the NAc Has an Antidepressant-like Effect The selective NK1 receptor antagonist RP67580, administered either systemically or directly into the NAc, significantly reduced escape latencies (ANOVA main drug effect, F4,54 = 3.13, p < 0.05), an effect shared by the subchronic administration of the antidepressant desipramine in this test. ∗p < 0.05 and ∗∗p < 0.01 from respective control conditions. Intracerebral injection, n = 7 mice per group; systemic drug injection, n = 17–18 group. Data are expressed as mean ± SEM. Neuron 2007 55, 289-300DOI: (10.1016/j.neuron.2007.06.033) Copyright © 2007 Elsevier Inc. Terms and Conditions

Figure 5 Viral Overexpression of ΔFosB in the vlPAG Opposes the Development of an IS-Induced Escape Deficit and Immobility in the Forced Swimming Test IS administration increased the percentage of escape failures (A) and average escape latency (B) in mice that received intra-vlPAG injections of the control HSV-LacZ vector (n = 19), similar to results seen in noninjected animals (see Figure 1). In contrast, HSV-ΔFosB (n = 18), as well as HSV-ΔJunD (n = 11), reduced the development of such escape deficits, while HSV-FosB (n = 12) had the opposite effect. ANOVA main effect of IS on escape latency, F3,68 = 12.88, p < 0.001, and percentage of escape failures, F3,68 = 12.36, p < 0.001, ∗p < 0.05, ∗∗∗p < 0.001 versus animals from the no-IS condition. None of these viral vectors altered escape performances in the absence of IS administration. Overexpression of ΔFosB also decreased immobility in the rat forced swimming test (C) (p < 0.01) while increasing time spent struggling (p < 0.05) and swimming (p < 0.05); n = 8–9 per group. On the other hand, HSV-ΔFosB did not alter sensitivity to foot shocks (D), general locomotion (E), motor coordination (F), or social interactions (not shown). Data are expressed as mean ± SEM. Neuron 2007 55, 289-300DOI: (10.1016/j.neuron.2007.06.033) Copyright © 2007 Elsevier Inc. Terms and Conditions

Figure 6 Viral Overexpression of ΔFosB in the vlPAG Example of HSV-ΔFosB infection in the vlPAG. The injections were aimed at the lateral margins of the DR in order to mimic the endogenous pattern of ΔFosB induction after IS (see Figure 1B and Figure S1). (A) Serotonergic neurons were immunolabeled with an antibody against tryptophan hydroxylase, and (B) the viral overexpression was revealed with an antibody against ΔFosB. The overlays at low (C) and high (D) magnification confirm the location of the infection at the lateral margins of the serotonergic cell groups (D). Dashed triangles indicate the position of the aqueduct on each picture. Neuron 2007 55, 289-300DOI: (10.1016/j.neuron.2007.06.033) Copyright © 2007 Elsevier Inc. Terms and Conditions